Package doBy

November 28, 2013

Version 4.5-10 Title doBy - Groupwise summary statistics, LSmeans, general linear contrasts, various utilities Author Sren Hjsgaard <sorenh@math.aau.dk> and Ulrich Halekoh <Ulrich.Halekoh@agrsci.dk> with contributions from Jim RobisonCox, Kevin Wright and Alessandro A. Leidi. Maintainer Sren Hjsgaard <sorenh@math.aau.dk> Description doBy originally contained facilities to 'do something' to data where data is partitioned 'by' some variables which dene groupings of data - hence the name doBy. doBy contains a variety of utilities including: 1) Facilities for groupwise computations of summary statistics and other facilities for working with grouped data. 2) LSmeans (least-squares means), general linear contrasts. 3) Miscellaneous other utilities. Encoding latin1 URL http://people.math.aau.dk/~sorenh/software/doBy/ ZipData no License GPL (>= 2) Depends R (>= 3.0), survival, MASS Imports Matrix, lme4 Suggests pbkrtest (>= 0.3-8), ggplot2, multcomp, geepack NeedsCompilation no Repository CRAN Date/Publication 2013-11-28 11:50:33 1

beets

R topics documented:
beets . . . . . . . . budworm . . . . . carcass . . . . . . . codstom . . . . . . createFunBy . . . . descStat . . . . . . dietox . . . . . . . doBy . . . . . . . . dose.LD50 . . . . . esticon . . . . . . . rstlastobs . . . . . lapplyBy . . . . . . linest . . . . . . . . lmBy . . . . . . . LSmeans . . . . . milkman . . . . . . NIRmilk . . . . . . orderBy . . . . . . parseGroupFormula recodeVar . . . . . renameCol . . . . . sampleBy . . . . . scaleBy . . . . . . splitBy . . . . . . . subSeq . . . . . . . subsetBy . . . . . . summaryBy . . . . timeSinceEvent . . transformBy . . . . which.maxn . . . . Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3 5 6 8 9 10 11 12 13 15 16 17 18 19 21 22 22 23 24 25 26 27 29 30 31 32 34 35 36 38

beets

Yield and sugar percentage in sugar beets from a split plot experiment.

Description Data is obtained from a split plot experiment. There are 3 blocks and in each of these the harvest time denes the "whole plot" and the sowing time denes the "split plot". Each plot was 25m2 and the yield is recorded in kg. See details for the experimental layout. Usage data(beets)

budworm Format The format is: chr "beets" Details Experimental plan Sowing times 1 2 3 4 5 1 2 4. 12. 21. 29. 18. 2. 21. april april april april may october october

Harvest times

Plot allocation: Block 1 Block 2 Block 3 +-----------|-----------|-----------+ Plot | 1 1 1 1 1 | 2 2 2 2 2 | 1 1 1 1 1 | 1-15 | 3 4 5 2 1 | 3 2 4 5 1 | 5 2 3 4 1 | |-----------|-----------|-----------| Plot | 2 2 2 2 2 | 1 1 1 1 1 | 2 2 2 2 2 | 16-30 | 2 1 5 4 3 | 4 1 3 2 5 | 1 4 3 2 5 | +-----------|-----------|-----------+ Examples
data(beets) ## maybe str(beets) ; plot(beets) ... beets$bh <- with(beets, interaction(block, harvest)) summary(aov(yield~block+sow+harvest+Error(bh), beets)) summary(aov(sugpct~block+sow+harvest+Error(bh), beets))

Harvest time Sowing time Harvest time Sowing time

budworm

Effect of Insecticide on survivial of tobacco budworms

Description Number of killed budworms after exposure to an insecticide. Usage data(budworm)

4 Format This data frame contains 12 rows and 4 columns: sex: sex of the budworm dose: dose of the insecticide trans-cypermethrin in g ndead: budworms killed in a trial ntotal: total number of budworms exposed per trial Details

budworm

Mortality of the moth tobacco budworm Heliothis virescens for 6 doses of the pyrethroid transcypermethrin differentiated with respect to sex. Source Collet, D. (1991) Modelling Binary Data, Chapman & Hall, London, Example 3.7 References Venables, W.N; Ripley, B.D.(1999) Modern Applied Statistics with S-Plus, Heidelberg, Springer, 3rd edition, chapter 7.2 Examples
data(budworm) ## function to caclulate the empirical logits empirical.logit<- function(nevent,ntotal) { y<-log ((nevent+0.5)/(ntotal-nevent+0.5)) y } ## plot the empirical logits against log-dose log.dose <- log(budworm$dose) emp.logit <- empirical.logit(budworm$ndead,budworm$ntotal) plot(log.dose,emp.logit,type= n ,xlab= log-dose ,ylab= emprirical logit ) title( budworm: emprirical logits of probability to die ) male <- budworm$sex== male female <- budworm$sex== female lines(log.dose[male],emp.logit[male],type= b ,lty=1,col=1) lines(log.dose[female],emp.logit[female],type= b ,lty=2,col=2) legend(0.5,2,legend=c( male , female ),lty=c(1,2),col=c(1,2))

carcass

carcass

Lean meat contents of 344 pig carcasses

Description Measurement of lean meat percentage of 344 pig carcasses together with auxillary information collected at three Danish slaughter houses Usage data(carcass) data(carcassall) Format carcassall: A data frame with 344 observations on the following 17 variables. weight Weight of carcass lengthc Length of carcass from back toe to head (when the carcass hangs in the back legs) lengthf Length of carcass from back toe to front leg (that is, to the shoulder) lengthp Length of carcass from back toe to the pelvic bone Fat02, Fat03, Fat11, Fat12, Fat13, Fat14, Fat16 Thickness of fat layer at different locations on the back of the carcass (FatXX refers to thickness at (or rather next to) rib no. XX. Notice that 02 is closest to the head Meat11, Meat12, Meat13 Thickness of meat layer at different locations on the back of the carcass, see description above LeanMeat Lean meat percentage determined by dissection slhouse Slaughter house; a factor with levels a b c sex Sex of the pig; a factor with a b c. Notice that it is no an error to have three levels; the third level refers to castrates carcass: Contains only the variables Fat11, Fat12, Fat13, Meat11, Meat12, Meat13, LeanMeat Source Busk, H., Olsen, E. V., Brndum, J. (1999) Determination of lean meat in pig carcasses with the Autofom classication system, Meat Science, 52, 307-314

codstom

codstom

Diet of Atlantic cod in the Gulf of St. Lawrence (Canada)

Description Stomach content data for Atlantic cod (Gadus morhua) in the Gulf of St.Lawrence, Eastern Canada. Note: many prey items were of no interest for this analysis and were regrouped into the "Other" category. Usage data(codstom) Format A data frame with 10000 observations on the following 10 variables. region a factor with levels SGSL NGSL representing the southern and northern Gulf of St. Lawrence, respectively ship.type a factor with levels 2 3 31 34 90 99 ship.id a factor with levels 11558 11712 136148 136885 136902 137325 151225 151935 99433 trip a factor with levels 10 11 12 179 1999 2 2001 20020808 3 4 5 6 7 8 88 9 95 set a numeric vector fish.id a numeric vector fish.length a numeric vector, length in mm prey.mass a numeric vector, mass of item in stomach, in g prey.type a factor with levels Ammodytes_sp Argis_dent Chion_opil Detritus Empty Eualus_fab Eualus_mac Gadus_mor Hyas_aran Hyas_coar Lebbeus_gro Lebbeus_pol Leptocl_mac Mallot_vil Megan_norv Ophiuroidea Other Paguridae Pandal_bor Pandal_mon Pasiph_mult Sabin_sept Sebastes_sp Them_abys Them_comp Them_lib Details Cod are collected either by contracted commerical shing vessels (ship.type 90 or 99) or by research vessels. Commercial vessels are identied by a unique ship.id. Either one research vessel or several commercial vessels conduct a survey (trip), during which a trawl, gillnets or hooked lines are set several times. Most trips are random stratied surveys (depth-based stratication). Each trip takes place within one of the regions. The trip label is only guaranteed to be unique within a region and the set label is only guaranteed to be unique within a trip. For each sh caught, the fish.length is recorded and the sh is allocated a fish.id, but the fish.id is only guaranteed to be unique within a set. A subset of the sh caught are selected for stomach analysis (stratied random selection according to sh length; unit of stratication is the

codstom

set for research surveys, the combination ship.id and stratum for surveys conducted by commercial vessels, although strata are not shown in codstom). The basic experimental unit in this data set is a cod stomach (one stomach per sh). Each stomach is uniquely identied by a combination of region, ship.type, ship.id, trip, set, and fish.id. For each prey item found in a stomach, the species and mass of the prey item are recorded, so there can be multiple observations per stomach. There may also be several prey items with the same prey.type in the one stomach (for example many prey.types have been recoded Other, which produced many instances of Other in the same stomach). If a stomach is empty, a single observation is recorded with prey.type Empty and a prey.mass of zero. Source Small subset from a larger dataset (more stomachs, more variables, more prey.types) collected by D. Chabot and M. Hanson, Fisheries & Oceans Canada (chabotd@dfo-mpo.gc.ca). Examples
data(codstom) str(codstom) # removes multiple occurences of same prey.type in stomachs codstom1 <- summaryBy(prey.mass ~ region+ship.type+ship.id+trip+set+fish.id+prey.type, data = codstom, id = ~fish.length, keep.names=TRUE, FUN = sum) # keeps a single line per stomach with the total mass of stomach content codstom2 <- summaryBy(prey.mass ~ region+ship.type+ship.id+trip+set+fish.id, data = codstom, id = ~fish.length, keep.names=TRUE, FUN = sum) # mean prey mass per stomach for each trip codstom3 <- summaryBy(prey.mass ~ region+ship.type+ship.id+trip, data = codstom2, keep.names=TRUE, FUN = mean) ## Not run: # wide version, one line per stomach, one column per prey type library(reshape) codstom4 <- melt(codstom, id = c(1:7, 9)) codstom5 <- cast(codstom4, region+ship.type+ship.id+trip+set+fish.id+fish.length ~ prey.type, sum) k <- length(names(codstom5)) prey_col <- 8:k out <- codstom5[,prey_col] out[is.na(out)] <- 0 codstom5[,prey_col] <- out codstom5$total.content <- rowSums(codstom5[, prey_col]) ## End(Not run)

createFunBy

createFunBy

A template function for creating groupwise functions

Description A template function for creating groupwise functions Usage formulaFunBy(formula, group, data, FUN, class = NULL, ...) xyFunBy(xy, group, data, FUN, class = NULL, ...) Arguments formula xy group data FUN class ... Value A function Note This function is a recent addition and has not been thoroughly tested. Please report bugs. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also parseGroupFormula A formula of the form y ~ x (which must be variable names in data). A character vector with one or two elements (which must be variable names in data). A right hand sided formula or a character vector dening the grouping of data A data frame The function to be applied The class to give the result of the returned value of the created function. Further arguments passed on to FUN

descStat Examples
## Example: Create a function for creating groupwise t-tests mydata <- data.frame(y=rnorm(32), x=rnorm(32), g1=factor(rep(c(1,2),each=16)), g2=factor(rep(c(1,2), each=8)), g3=factor(rep(c(1,2),each=4))) t.testBy1 <- function(formula, group, data, ...){ formulaFunBy(formula, group, data, FUN=t.test, class="t.testBy1", ...) } t.testBy2 <- function(formula, group, data, ...){ xyFunBy(formula, group, data, FUN=t.test, class="t.testBy1", ...) } t.testBy1(y~g1, ~g2+g3, data=mydata) t.testBy2(y~x, ~g2+g3, data=mydata)

descStat

Computing simple descriptive statistics of a numeric vector.

Description Computing simple descriptive statistics of a numeric vector not unlike what proc means of SAS does Usage descStat(x, na.rm = TRUE) Arguments x na.rm Value A vector with named elements. Author(s) Gregor Gorjanc; gregor.gorjanc <at> bf.uni-lj.si A numeric vector Should missing values be removed

10 See Also summaryBy Examples

x <- c(1,2,3,4,NA,NaN) descStat(x)

dietox

dietox

Growth curves of pigs in a 3x3 factorial experiment

Description The dietox data frame has 861 rows and 7 columns. Data contains weight of slaughter pigs measured weekly for 12 weeks. Data also contains the startweight (i.e. the weight at week 1). The treatments are 3 different levels of Evit = vitamin E (dose: 0, 100, 200 mg dl-alpha-tocopheryl acetat /kg feed) in combination with 3 different levels of Cu=copper (dose: 0, 35, 175 mg/kg feed) in the feed. The cumulated feed intake is also recorded. The pigs are littermates. Usage data(dietox) Format This data frame contains the following columns: Weight Weight Feed Cumulated feed intake Time Time (in weeks) in the experiment Pig Id of each pig Evit Vitamin E dose Cu Copper dose Start Start weight in experiment, i.e. weight at week 1. Litter Id of litter of each pig Source Lauridsen, C., Hjsgaard, S.,Srensen, M.T. C. (1999) Inuence of Dietary Rapeseed Oli, Vitamin E, and Copper on Performance and Antioxidant and Oxidative Status of Pigs. J. Anim. Sci.77:906916

doBy Examples
data(dietox) str(dietox) ; plot(dietox)

11

doBy

Various utilitie. Functions for creating groupwise calculations etc.; calculation of least-squares means; miscellaneous utilities

Description The core doBy functions were developed to make it easy to split data into groups (dened by the levels of a set of factors) and performing some actions on each of these groups. Thus, these functions mimic what can be achieved using the BY statement in various SAS procedures. In addition hereto the doBy package containts various other utilities. Details Functions summaryBy, splitBy, orderBy, sampleBy, transformBy etc. are the doBy functions. linest() calculates linear estimates based on a matrix for various model objects. The esticon() function has a similar functionality, but it will be removed at some point of time. LSmeans() and LSmatrix() are used in connection with calculating leastsquares means. There are various other utility functions in the package. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also summaryBy, orderBy, transformBy, splitBy, sampleBy Examples
data(dietox) summaryBy(Weight+Feed~Evit+Cu+Time, data=dietox, FUN=c(mean,var), na.rm=TRUE, use="pair") orderBy(~Time+Evit, data=dietox) splitBy(formula = ~Evit+Cu, data = dietox) sampleBy(formula = ~Evit+Cu, frac=.1, data = dietox)

12

dose.LD50

dose.LD50

Calculate LD50

Description Calculate the LD50 (the dose at which 50 pct of the subjects die) for a model of the form logit(p)=beta1 x1 + ... + betap xp + gamma d where none of the explanatory variables x1 ... xp contains the dose d. Usage dose.LD50(x, lambda) Arguments x lambda Details lambda contains an NA at the entry corresponding to dose d. The other entries of lambda must be the values of the covariates x1 ... xp at which the ld50 is to be calculated. Value A data frame Author(s) Sren Hjsgaard Examples
data(budworm) m1 <- glm(ndead/20 ~ sex + log(dose), data=budworm, weight=ntotal, family=binomial) coef(m1) dose.LD50(m1,c(1,1,NA)) dose.LD50(m1,c(1,0,NA))

A glm object (for logistic regression) A vector of the same length as the number of parameters in x.

esticon

13

esticon

Contrasts for lm, glm, lme, and geeglm objects

Description Computes linear functions (i.e. weighted sums) of the estimated regression parameters. Can also test the hypothesis, that such a function is equal to a specic value. Usage esticon(obj, cm, beta0, conf.int = TRUE, level=0.95, joint.test = FALSE,...) Arguments obj cm Regression object (of type lm, glm, lme, geeglm) Matrix specifying linear functions of the regresson parameters (one linear function per row). The number of columns must match the number of tted regression parameters in the model. See details below. A vector of numbers TRUE The condence level Logical value. If TRUE a joint Wald test for the hypothesis L beta=beta0 is made. Default is that the row-wise tests are made, i.e. (L beta)i=beta0i. If joint.test is TRUE, then no condence inteval etc. is calculated. Additional arguments; currently not used.

beta0 conf.int level joint.test

... Details

Let the estimated parameters of the model be latex A linear function of the estimates is of the form latex where latex is specied by the user. The esticon function calculates c, its standard error and by default also a 95 pct condence interval. It is sometimes of interest to test the hypothesis latex for some value latex given by the user. A test is provided for the hypothesis latex but other values of latex can be specied. In general, one can specify r such linear functions at one time by specying cm to be an latex matrix where each row consists of p numbers latex. Default is then that latex is a p vector of 0s but other values can be given. It is possible to test simulatneously that all specied linear functions are equal to the corresponding values in latex. For computing contrasts among levels of a single factor, contrast.lm may be more convenient.

14 Value

esticon

Returns a matrix with one row per linear function. Columns contain estimated coefcients, standard errors, t values, degrees of freedom, two-sided p-values, and the lower and upper endpoints of the 1-alpha condence intervals. Note esticon works on geese/geeglm objects from the geepack package (for Generalized Estimating Equations), on lm and glm objects, and on gls objects. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> Examples
data(iris) lm1 <- lm(Sepal.Length~Sepal.Width+Species+Sepal.Width:Species, data=iris) ## Note that the setosa parameters are set to zero coef(lm1) ## Estimate the intercept for versicolor lambda1 <- c(1,0,1,0,0,0) esticon(lm1,lambda1) ## Estimate the difference between versicolor and virgica intercept ## and test if the difference is 1 lambda2 <- c(0,1,-1,0,0,0) esticon(lm1,lambda2,beta0=1) ## Do both estimates at one time esticon(lm1,rbind(lambda1,lambda2),beta0=c(0,1)) ## Make a combined test for that the difference between versicolor and virgica intercept ## and difference between versicolor and virginica slope is zero: lambda3 <- c(0,0,0,0,1,-1) esticon(lm1,rbind(lambda2,lambda3),joint.test=TRUE) # Example using esticon on coxph objects (thanks to Alessandro A. Leidi). # Using dataset veteran in the survival package # from the Veterans Administration Lung Cancer study library(survival); data(veteran) sapply(veteran,class) levels(veteran$celltype) attach(veteran) veteran.s<-Surv(time,status) coxmod<-coxph(veteran.s~age+celltype+trt,method= breslow ) summary(coxmod) # compare a subject 50 years old with celltype 1

rstlastobs
# to a subject 70 years old with celltype 2 # both subjects on the same treatment AvB<-c(-20,-1,0,0,0) # compare a subject 40 years old with celltype 2 on treat=0 # to a subject 35 years old with celltype 3 on treat=1 CvB<-c(5,1,-1,0,-1) esti<-esticon(coxmod,rbind(AvB,CvB)) esti exp(esti[,c(2,7,8)])

15

firstlastobs

Locate the index of the rst/last unique value

Description Locate the index of the rst/last unique value in i) a vector or of a variable in a data frame. Usage ## S3 method for class formula firstobs(formula, data=parent.frame(), ...) ## S3 method for class formula lastobs(formula, data=parent.frame(), ...) firstobs(x, ...) lastobs(x, ...) Arguments x formula data ... Details If writing ~a+b+c as formula, then only a is considered. Value A vector. A vector A formula (only the rst term is used, see details). A data frame Currently not used

16 Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> Examples

x <- c(rep(1,5),rep(2,3),rep(3,7),rep(1,4)) firstobs(x) lastobs(x) data(dietox) firstobs(~Pig, data=dietox) lastobs(~Pig, data=dietox)

lapplyBy

lapplyBy

Formula based version of lapply

Description This function is a wrapper for calling lapply on the list resulting from rst calling splitBy. Usage lapplyBy(formula, data = parent.frame(), FUN) Arguments formula data FUN A formula describing how data should be split A dataframe A function to be applied to each element in the splitted list, see Examples below.

Value A list. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also orderBy, summaryBy, transformBy, splitBy,

linest Examples
data(dietox) ## Calculate weekwise feed efficiency = weight gain / feed intake dietox <- orderBy(~Pig+Time, data=dietox) v<-lapplyBy(~Pig, data=dietox, function(d) c(NA, diff(d$Weight)/diff(d$Feed))) dietox$FE <- unlist(v) ## Technically this is the same as dietox <- orderBy(~Pig+Time, data=dietox) wdata <- splitBy(~Pig, data=dietox) v <- lapply(wdata, function(d) c(NA, diff(d$Weight)/diff(d$Feed))) dietox$FE <- unlist(v)

17

linest

Compute linear estimates

Description Compute linear estimates for a range of models. One example of linear estimates is population means (also known as LSMEANS). Usage linest(object, K, level=0.95, ...) Arguments object K Model object A contrast matrix. If given then the specications in effect and at are ignored. Otherwise, the matrix is generated according to the specications in effect and at. The level of the (asymptotic) condence interval. Additional arguments; currently not used.

level ... Value

A dataframe with results from computing the contrasts. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also LSmeans LSmatrix

18 Examples
## Make balanced dataset dat.bal <- expand.grid(list(AA=factor(1:2), BB=factor(1:3), CC=factor(1:3))) dat.bal$y <- rnorm(nrow(dat.bal)) ## Make unbalanced dataset # BB is nested within CC so BB=1 is only found when CC=1 # and BB=2,3 are found in each CC=2,3,4 dat.nst <- dat.bal dat.nst$CC <-factor(c(1,1,2,2,2,2,1,1,3,3,3,3,1,1,4,4,4,4)) mod.bal mod.nst <- lm(y ~ AA + BB*CC, data=dat.bal) <- lm(y ~ AA + BB : CC, data=dat.nst)

lmBy

K <- LSmatrix(mod.nst, effect=c("BB","CC")) linest( mod.nst, K )

lmBy

List of lm objects with a common model

Description The data is split into strata according to the levels of the grouping factors and individual lm ts are obtained for each stratum. Usage lmBy(formula, data, id = NULL, ...) Arguments formula A linear model formula object of the form y ~ x1+...+xn | g1+...gm. In the formula object, y represents the response, x1,...,xn the covariates, and the grouping factors specifying the partitioning of the data according to which different lm ts should be performed. A dataframe A formula describing variables from data which are to be available also in the output. Additional arguments passed on to lm().

data id ... Value A list of lm ts. Note

This is a recent addition to the package; please report bugs.

LSmeans Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> Examples

bb <- lmBy(1/uptake~log(conc)|Treatment, data=CO2) coef(bb) fitted(bb) residuals(bb) summary(bb) coef(summary(bb)) coef(summary(bb), simplify=TRUE)

19

LSmeans

Compute linear estimates

Description Compute linear estimates for a range of models. One example of linear estimates is population means (also known as LSMEANS). Usage LSmeans(object, effect = NULL, at = NULL, level=0.95, ...) Arguments object effect at level ... Value A dataframe with results from computing the contrasts. Model object A vector of variables. For each conguration of these the estimate will be calculated. A list of values of covariates (including levels of some factors) to be used in the calculations The level of the (asymptotic) condence interval. Additional arguments; currently not used.

20 Note

LSmeans

The LSmeans method is a recent addition to the package, and it will eventually replace the popMeans method. Please report unexpected behaviour. Some of the code has been adapted from the lsmeans package. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also LSmatrix linest Examples
## Make balanced dataset dat.bal <- expand.grid(list(AA=factor(1:2), BB=factor(1:3), CC=factor(1:3))) dat.bal$y <- rnorm(nrow(dat.bal)) ## Make unbalanced dataset # BB is nested within CC so BB=1 is only found when CC=1 # and BB=2,3 are found in each CC=2,3,4 dat.nst <- dat.bal dat.nst$CC <-factor(c(1,1,2,2,2,2,1,1,3,3,3,3,1,1,4,4,4,4)) mod.bal mod.nst <- lm(y ~ AA + BB*CC, data=dat.bal) <- lm(y ~ AA + BB : CC, data=dat.nst)

LSmeans(mod.bal, effect=c("BB", "CC")) LSmeans(mod.nst, effect=c("BB", "CC")) LSmeans(mod.nst, at=list(BB=1, CC=1)) LSmeans(mod.nst, at=list(BB=1, CC=2)) ## Above: NA s are correct; not an estimable function if( require( lme4 )){ warp.mm <- lmer(breaks ~ -1 + tension + (1|wool), data=warpbreaks) class(warp.mm) fixef(warp.mm) coef(summary(warp.mm)) vcov(warp.mm) if (require(pbkrtest )) vcovAdj(warp.mm) }

milkman

21

milkman

Milk yield data for manually milked cows.

Description Milk yield data for cows milked manually twice a day (morning and evening). Usage data(milkman) Format A data frame with 161836 observations on the following 12 variables. cowno a numeric vector; cow identication lactno a numeric vector; lactation number ampm a numeric vector; milking time: 1: morning; 2: evening dfc a numeric vector; days from calving my a numeric vector; milk yield (kg) fatpct a numeric vector; fat percentage protpct a numeric vector; protein percentage lactpct a numeric vector; lactose percentage scc a numeric vector; somatic cell counts race a factor with levels RDM Holstein Jersey ecmy a numeric vector; energy corrected milk cowlact Combination of cowno and lactno; necessary because the same cow may appear more than once in the dataset (in different lactations) Details There are data for 222 cows. Some cows appear more than once in the dataset (in different lactations) and there are 288 different lactations. Examples
data(milkman) ## maybe str(milkman) ; plot(milkman) ...

22

orderBy

NIRmilk

NIRmilk

Description Near infra red light (NIR) measurments are made at 152 wavelengths on 17 milk samples. While milk runs through a glass tube, infra red light is sent through the tube and the amount of light passing though the tube is measured at different wavelengths. Each milk sample was additionally analysed for fat, lactose, protein and drymatter. Usage data(NIRmilk) Format This data frame contains 18 rows and 158 columns. The rst column is the sample number. The columns Xwww contains the infra red light amount at wavelength www. The response variables are fat, protein, lactose and dm (drymatter). Details PCA regression Examples
data(NIRmilk)

orderBy

Ordering (sorting) rows of a data frame

Description Ordering (sorting) rows of a data frame by the certain variables in the data frame. This function is essentially a wrapper for the order() function - the important difference being that variables to order by can be given by a model formula. Usage orderBy(formula, data) Arguments formula data The right hand side of a formula A data frame

parseGroupFormula Details

23

The sign of the terms in the formula determines whether sorting should be ascending or decreasing; see examples below Value The ordered data frame Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> and Kevin Wright See Also summaryBy, transformBy, splitBy, lapplyBy, Examples
data(dietox) orderBy(~Time+Evit, data=dietox) ## Sort decreasingly by Time orderBy(~-Time+Evit, data=dietox)

parseGroupFormula

Extract components from a formula with "conditioning bar"

Description Extract components from a formula with the form y~x1+...xn|g1+...gm Usage parseGroupFormula(form) Arguments form Value If the formula is y~x1+x2|g1+g2 the result is model groups groupFormula y~x1+x2 g1+g2 ~g1+g2 A formula of the form y~x1+...xn|g1+...gm

24 Note

recodeVar

This function is a recent addition and has not been thoroughly tested. Please report bugs. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> Examples
gf<-parseGroupFormula(y~x1+x2|g1+g2)

recodeVar

Recode values of a vector

Description Recodes a vector with values, say 1,2 to a variable with values, say a, b Usage recodeVar(x, src, tgt, default=NULL, keep.na=TRUE) Arguments x src tgt default keep.na Value A vector Warning Care should be taken if x is a factor. A safe approach may be to convert x to a character vector using as.character. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> A vector; the variable to be recoded The source values: a subset of the present values of x The target values: the corresponding new values of x Default target value for those values of x not listed in src. When default=NULL, values of x which are not given in src will be kept in the output. If TRUE then NAs in x will be retained in the output

renameCol See Also cut, factor, recodeVar Examples

x <- c("dec","jan","feb","mar","apr","may") src1 <- list(c("dec","jan","feb"), c("mar","apr","may")) tgt1 <- list("winter","spring") recodeVar(x,src=src1,tgt=tgt1) #[1] "winter" "winter" "winter" "spring" "spring" "spring" x <- c(rep(1:3,3)) #[1] 1 2 3 1 2 3 1 2 3 ## Simple usage: recodeVar(x, src=c(1,2), tgt=c("A","B")) #[1] "A" "B" NA "A" "B" NA "A" "B" NA ## Here we need to use lists recodeVar(x, src=list(c(1,2)), tgt=list("A")) #[1] "A" "A" NA "A" "A" NA "A" "A" NA recodeVar(x, src=list(c(1,2)), tgt=list("A"), default="L") #[1] "A" "A" "L" "A" "A" "L" "A" "A" "L" recodeVar(x, src=list(c(1,2),3), tgt=list("A","B"), default="L") #[1] "A" "A" "B" "A" "A" "B" "A" "A" "B" ## Dealing with NA s in x x<-c(NA,rep(1:3,3),NA) #[1] NA 1 2 3 1 2 3 1 2 3 NA recodeVar(x, src=list(c(1,2)), tgt=list("A")) #[1] NA "A" "A" NA "A" "A" NA "A" "A" NA NA recodeVar(x, src=list(c(1,2)), tgt=list("A"), default="L") #[1] NA "A" "A" "L" "A" "A" "L" "A" "A" "L" NA recodeVar(x, src=list(c(1,2)), tgt=list("A"), default="L", keep.na=FALSE) #[1] "L" "A" "A" "L" "A" "A" "L" "A" "A" "L" "L" x <- c("no", "yes", "not registered", "no", "yes", "no answer") recodeVar(x, src = c("no", "yes"), tgt = c("0", "1"), default = NA)

25

renameCol

Rename columns in a matrix or a dataframe.

Description Rename columns in a matrix or a dataframe. Usage renameCol(indata, src, tgt)

26 Arguments indata src tgt Value A dataframe if indata is a dataframe; a matrix in indata is a matrix. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also rename.vars Examples
renameCol(CO2, 1:2, c("kk","ll")) renameCol(CO2, c("Plant","Type"), c("kk","ll")) # These fail - as they should: # renameCol(CO2, c("Plant","Type","conc"), c("kk","ll")) # renameCol(CO2, c("Plant","Type","Plant"), c("kk","ll"))

sampleBy

A dataframe or a matrix Source: Vector of names of columns in indata to be renamed. Can also be a vector of column numbers. Target: Vector with corresponding new names in the output.

sampleBy

Sampling from a data frame

Description A data frame is split according to some variables in a formula, and a sample of a certain fraction of each is drawn. Usage sampleBy(formula, frac = 0.1, replace=FALSE, data = parent.frame(),systematic=FALSE) Arguments formula frac replace data systematic A formula dening the grouping of the data frame The part of data to be sampled. Is the sampling with replacement A data frame Should sampling be systematic.

scaleBy Details

27

If systematic=FALSE (default) then frac gives the fraction of data sampled. If systematic=TRUE and frac=.2 then every 1/.2 i.e. every 5th observation is taken out. Value A data frame Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also orderBy, summaryBy, transformBy, splitBy, Examples
data(dietox) sampleBy(formula = ~Evit+Cu, frac=.1, data = dietox)

scaleBy

Groupwise scaling and centering of numeric columns in a dataframe

Description Groupwise scaling and centering of numeric columns in a dataframe. Obtained by rst splitting a dataframe and then calling scale on each stratum. Usage scaleBy(formula, data, center = TRUE, scale = TRUE, details=0) Arguments formula data center scale details Either a two-sided formula or a list. A dot (.) is allowed on both left and right hand side of formula. See details for the meaning of this. A dataframe If TRUE then data is centered to have mean zero If TRUE then data is scaled to have variance one If larger than zero then information about grouping etc. is printed.

28 Details

scaleBy

A typical formula is y1+y2~f1+f2 where y1 and y2 are numeric variables and f1 and f2 can be of any type. For each cross-combination of the values of f1 and f2, the variables (y1,y2) are centered/scaled. It is valid to write .~f1+f2. In this case the variables to be centered/scaled are taken to be all numeric variables in the dataframe except those that a listed on the right hand side of the formula. It is valid to write y1+y2~.. In this case the stratication is taken to be by all non-numeric variables. If there are no non-numeric variables, then no stratication is made and a global centering/scaling is made. It is valid to write .~.. In this case the variables to be centered/scaled are taken to be all numeric variables in the dataframe. The stratication is made by all non-numeric variables. If there are no non-numeric variables, then no stratication is made and a global centering/scaling is made. Value A dataframe with the same columns as the input dataframe, but the scaled / centered values are put into the relevant columns. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also lapplyBy, orderBy, splitBy, summaryBy, transformBy, Examples
## The following forms are equivalent: scaleBy(conc+rate ~ state, data=Puromycin) scaleBy(list(c("conc","rate"), "state"), data=Puromycin) scaleBy(list(c("."), "."), data=Puromycin) scaleBy(.~., data=Puromycin) ## The same results can be obtained from lapply(splitBy(~state, data=Puromycin), function(.dd) scale(.dd[,sapply(Puromycin,class)=="numeric"])) ## The pig growth data data(dietox) dietox

# "Remove the effect of time" by centering data within each time point. dietox2 <- scaleBy(Weight~Time, data=dietox, scale=FALSE) ## Not run: library(lattice)

splitBy
xyplot(Weight~Time|Evit+Cu, groups=Pig, data=dietox) xyplot(Weight~Time|Evit+Cu, groups=Pig, data=dietox2) ## End(Not run)

29

splitBy

Split a data frame

Description Split a dataframe according to the levels of variables in the dataframe. The variables to split by can be given as a formula or as a character vector. Usage splitBy(formula, data = parent.frame(), drop=TRUE)

Arguments formula data drop ... Value A list of dataframes. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also orderBy, summaryBy, transformBy, lapplyBy, scaleBy Examples
data(dietox, package="doBy") splitBy(formula = ~Evit+Cu, data = dietox) splitBy(formula = c("Evit","Cu"), data = dietox) splitBy(~Month, data=airquality) splitBy("Month", data=airquality)

The right hand side of a formula (or a character vector) A data frame Logical indicating if levels that do not occur should be dropped Additional arguments, currently not used.

30

subSeq

subSeq

Find sub-sequences of identical elements in a vector.

Description Find sub-sequences of identical elements in a vector. Usage subSeq(x, item = NULL) Arguments x item Value A dataframe. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also rle Examples
x <- c(1,1,1,0,0,1,1,1,2,2,2,1,2,2,2,3) (ans <- subSeq(x)) ans$value # Notice: Same results below subSeq(x,item=1) subSeq(x,item="1") x <- as.character(c(1,1,1,0,0,1,1,1,2,2,2,1,2,2,2,3)) (ans<-subSeq(x)) ans$value # Notice: Same results below subSeq(x,item="1") subSeq(x,item=1)

An atomic vector. Optionally a specic value to look for in x.

subsetBy

31

subsetBy

Finds subsets of a dataframe which is split by variables in a formula.

Description A data frame is split by a formula into groups. Then subsets are found within each group, and the result is collected into a data frame. Usage subsetBy(formula, subset, data = parent.frame(), select, drop=FALSE, join=TRUE, ... ) Arguments formula subset data select drop join ... Value A data frame. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also See Also splitBy Examples
data(dietox) subsetBy(~Evit, Weight < mean(Weight), data=dietox)

A formula to split by logical expression indicating elements or rows to keep: missing values are taken as false. A data frame expression, indicating columns to select from a data frame. passed on to [ indexing operator. If FALSE the result is a list of data frames (as dened by formula); if TRUE one data frame is returned. further arguments to be passed to or from other methods.

32

summaryBy

summaryBy

Function to calculate groupwise summary statistics

Description Function to calculate groupwise summary statistics, much like the summary procedure of SAS Usage summaryBy(formula, data = parent.frame(), id = NULL, FUN = mean, keep.names=FALSE, p2d=FALSE, order=TRUE, full.dimension=FALSE, var.names=NULL, fun.names=NULL, ...) Arguments formula data id FUN keep.names p2d order A formula object, see examples below A data frame A formula specifying variables which data are not grouped by but which should appear in the output. See examples below. A list of functions to be applied, see examples below. If TRUE and if there is only ONE function in FUN, then the variables in the output will have the same name as the variables in the input, see examples. Should parentheses in output variable names be replaced by dots? Should the resulting dataframe be ordered according to the variables on the right hand side of the formula? (using orderBy

full.dimension If TRUE then rows of summary statistics are repeated such that the result will have the same number of rows as the input dataset. var.names fun.names ... Details Extra arguments (...) are passed onto the functions in FUN. Hence care must be taken that all functions in FUN accept these arguments - OR one can explicitly write a functions which get around this. This can particularly be an issue in connection with handling NAs. See examples below. Some code for this function has been suggested by Jim Robison-Cox. Value A data frame Option for user to specify the names of the variables on the left hand side. Option for user to specify function names to apply to the variables on the left hand side. Additional arguments to FUN. This could for example be NA actions.

summaryBy Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also ave, descStat, lapplyBy, orderBy, scaleBy, splitBy, transformBy, Examples
data(dietox) dietox12 <- subset(dietox,Time==12) summaryBy(Weight+Feed~Evit+Cu, data=dietox12, FUN=c(mean,var,length)) summaryBy(list(c("Weight","Feed"), c("Evit","Cu")), data=dietox12, FUN=c(mean,var,length)) summaryBy(Weight+Feed~Evit+Cu+Time, data=subset(dietox,Time>1), FUN=c(mean,var,length)) ## Calculations on transformed data: summaryBy(log(Weight)+Feed~Evit+Cu, data=dietox12) ## Calculations on all numerical variables (not mentioned elsewhere): summaryBy(.~Evit+Cu, id=~Litter, FUN=mean) ## ## ## ## ## data=dietox12,

33

There are missing values in the airquality data, so we remove these before calculating mean and variance with na.rm=TRUE . However the length function does not accept any such argument. Hence we get around this by defining our own summary function in which length is not supplied with this argument while mean and var are:

sumfun <- function(x, ...){ c(m=mean(x, ...), v=var(x, ...), l=length(x)) } summaryBy(Ozone+Solar.R~Month, data=airquality, FUN=sumfun, na.rm=TRUE) ## Using . on the right hand side of a formula means to stratify by ## all variables not used elsewhere: data(warpbreaks) summaryBy(breaks ~ wool+tension, warpbreaks) summaryBy(breaks ~., warpbreaks) summaryBy(.~ wool+tension, warpbreaks) ## Keep the names of the variables (works only if FUN only returns one

34
## value): summaryBy(Ozone+Wind~Month, data=airquality,FUN=c(mean),na.rm=TRUE, keep.names=TRUE) ## Using full.dimension=TRUE

timeSinceEvent

## Consider: summaryBy(breaks~wool, data=warpbreaks) ## Rows of result are replicated below summaryBy(breaks~wool, data=warpbreaks, full.dimension=TRUE) ## Notice: Previous result is effectively the same as with(warpbreaks, ave(breaks, wool)) ## A possible application of full.dimension=TRUE is if we want to ## standardize (center and scale) data within groups: ss <- summaryBy(breaks~wool, data=warpbreaks, full.dimension=TRUE, FUN=c(mean,sd)) (warpbreaks$breaks-ss$breaks.mean)/ss$breaks.sd

timeSinceEvent

Calculate "time since event" in a vector.

Description Events are coded as 1 in numeric vector (and non-events are coded with values different from 1). timeSinceEvent will give the time since event (with and without sign). In a logical vector, events are coded as TRUE and all non-events as FALSE. Usage timeSinceEvent(yvar, tvar = seq_along(yvar)) Arguments yvar tvar Value A dataframe with columns yvar, tvar, abs.tse (absolute time since nearest event), sign.tse (signed time since nearest event) and run (indicator of the time window around each event). Note NAs in yvar are converted to zeros. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> A numerical or logical vector specifying the events An optional vector specifying time

transformBy See Also subSeq, rle Examples

## Events: yvar <- c(0,0,0,1,0,0,0,0,0,1,0,0,0,0,0,1,0,1,0,0,0,0,0,0,0,0,1,1,0,0,0,0,0) ## Plot results: tse<- timeSinceEvent(yvar) plot(sign.tse~tvar, data=tse, type="b") grid() rug(tse$tvar[tse$yvar==1], col=4,lwd=4) points(scale(tse$run), col=tse$run,lwd=2) lines(abs.tse+.2~tvar, data=tse, type="b",col=3) ## Find times for which time since an event is at most 1: tse$tvar[tse$abs<=1] yvar <- c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ) tvar <- c(207, 208, 208, 208, 209, 209, 209, 209, 210, 210, 211, 211, 211, 212, 213, 213, 214, 214, 215, 216, 216, 216, 216, 217, 217, 217, 218, 218, 219, 219, 219, 219, 220, 220, 221, 221, 221, 221, 222, 222, 222) timeSinceEvent(yvar, tvar)

35

transformBy

Function to make groupwise transformations

Description Function to make groupwise transformations of data by applying the transform function to subsets of data. Usage transformBy(formula, data, ...) Arguments formula data ... A formula with only a right hand side, see examples below A data frame Further arguments of the form tag=value

36 Details

which.maxn

The ... arguments are tagged vector expressions, which are evaluated in the data frame data. The tags are matched against names(data), and for those that match, the value replace the corresponding variable in data, and the others are appended to data. Value The modied value of the dataframe data. Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also orderBy, summaryBy, splitBy, doby.xtabs, Examples
data(dietox) transformBy(~Pig, data=dietox, minW=min(Weight), maxW=max(Weight), gain=sum(range(Weight)*c(-1,1)))

which.maxn

Where are the n largest or n smallest elements in a numeric vector ?

Description Determines the locations, i.e., indices of the n largest or n smallest elements of a numeric vector. Usage which.maxn(x, n = 1) which.minn(x, n = 1) Arguments x n Value A vector of length at most n with the indices of the n largest / smaller elements. NAs are discared and that can cause the vector to be smaller than n. numeric vector integer >= 1

which.maxn Author(s) Sren Hjsgaard, <sorenh@math.aau.dk> See Also which.max, which.min Examples
x <- c(1:4,0:5,11,NA,NA) ii <- which.minn(x,5) x <- c(1,rep(NA,10),2) ii <- which.minn(x,5)

37

Index
Topic datasets beets, 2 budworm, 3 carcass, 5 codstom, 6 dietox, 10 milkman, 21 NIRmilk, 22 Topic models dose.LD50, 12 lmBy, 18 Topic univar summaryBy, 32 transformBy, 35 Topic utilities createFunBy, 8 descStat, 9 doBy, 11 esticon, 13 firstlastobs, 15 lapplyBy, 16 linest, 17 LSmeans, 19 orderBy, 22 parseGroupFormula, 23 recodeVar, 24 sampleBy, 26 scaleBy, 27 splitBy, 29 subSeq, 30 subsetBy, 31 timeSinceEvent, 34 which.maxn, 36 Topic utitlities renameCol, 25 ave, 33 beets, 2 budworm, 3 38 carcass, 5 carcassall (carcass), 5 codstom, 6 coef.lmBy (lmBy), 18 coef.summary.lmBy (lmBy), 18 createFunBy, 8 cut, 25 descStat, 9, 33 dietox, 10 doBy, 11 doby.xtabs, 36 dose.LD50, 12 esticon, 13 factor, 25 firstlastobs, 15 firstobs (firstlastobs), 15 fitted.lmBy (lmBy), 18 formulaFunBy (createFunBy), 8 getBy (lmBy), 18 lapplyBy, 16, 23, 28, 29, 33 lastobs (firstlastobs), 15 linest, 17, 20 lmBy, 18 LSmatrix, 17, 20 LSmeans, 17, 19 milkman, 21 NIRmilk, 22 orderBy, 11, 16, 22, 2729, 32, 33, 36 parseGroupFormula, 8, 23 recodeVar, 24, 25 rename.vars, 26

INDEX renameCol, 25 residuals.lmBy (lmBy), 18 rle, 30, 35 sampleBy, 11, 26 scaleBy, 27, 29, 33 splitBy, 11, 16, 23, 27, 28, 29, 31, 33, 36 subSeq, 30, 35 subsetBy, 31 summary.lmBy (lmBy), 18 summaryBy, 10, 11, 16, 23, 2729, 32, 36 timeSinceEvent, 34 transformBy, 11, 16, 23, 2729, 33, 35 which.max, 37 which.maxn, 36 which.min, 37 which.minn (which.maxn), 36 xyFunBy (createFunBy), 8

39